Research map / Federated learning and privacy / Personalized and heterogeneous FL
Fed-Duet: Dual Expert-Orchestrated Framework for Continual Federated Vision-Language Learning
Tao GUO, Junwei Chen, Laizhong Cui
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add parametric (adapter) experts to the semantic-only pathway (ablation) | Fed-Duet · Table 6 ablation | average accuracy: 64.34 -> 77.96 | Improved |
| add semantic (prompt) experts to the parametric-only pathway (ablation) | Fed-Duet · Table 6 ablation | average accuracy: 70.64 -> 77.96 | Improved |
| add routing consistency loss Lcross modal (ablation) | Fed-Duet · Table 6 ablation | average accuracy: 77.96 -> 79.09 | Improved |
| add expert stability loss Lstability (ablation) | Fed-Duet · Table 6 ablation | forgetting: 9.22 -> 8.02 | Improved |
| add both auxiliary losses to Base (ablation) | Fed-Duet · Table 6 ablation | average accuracy: 77.96 -> 80.43 | Improved |
| train parametric experts first instead of semantic experts first (ablation) | Fed-Duet · CIFAR-100 | accuracy: 85.53 -> 86.21 | Improved |
| train parametric then semantic experts instead of joint training (ablation) | Fed-Duet · CIFAR-100 | accuracy: 86.02 -> 86.21 | Improved |
| move stage transition from R/4 to R/2 (ablation) | Fed-Duet · CIFAR-100 and Tiny-ImageNet average | accuracy: 81.94 -> 83.47 | Improved |
| move stage transition from R/2 to 3R/4 (ablation) | Fed-Duet · CIFAR-100 and Tiny-ImageNet average | accuracy: 83.47 -> 83.43 | No clear effect |
| initialize prompt pool with K-means centroids instead of random (ablation) | Fed-Duet · CIFAR-100 | accuracy: 84.61 -> 85.19 | Improved |
| use gating network instead of random prompt selection (ablation) | Fed-Duet · CIFAR-100 and Tiny-ImageNet average | accuracy: 82.22 -> 83.47 | Improved |
| add local expert logits to the shared expert logits (lambda 0.00 to 0.50) (ablation) | Fed-Duet · CIFAR-100 | accuracy: 31.05 -> 86.21 | Improved |
Closest papers by meaning
- Mixture of Experts Made Personalized: Federated Prompt Learning for Vision-Language Models (ICLR 2025 · similarity 0.89)
- F^3OCUS - Federated Finetuning of Vision-Language Foundation Models with Optimal Client Layer Updating Strategy via Multi-objective Meta-Heuristics (CVPR 2025 · similarity 0.88)
- Federated Continual Learning Goes Online: Uncertainty-Aware Memory Management for Vision Tasks and Beyond (ICLR 2025 · similarity 0.88)
- FedPHA: Federated Prompt Learning for Heterogeneous Client Adaptation (ICML 2025 · similarity 0.88)
- Federated Disentangled Tuning with Textual Prior Decoupling and Visual Dynamic Adaptation (ICML 2025 · similarity 0.87)
- Global Prompt Refinement with Non-Interfering Attention Masking for One-Shot Federated Learning (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
